Aalborg University Department of Computer Science Selma Lagerlöfs Vej 300 9220 Aalborg Telephone:(45)96358080 http://www.cs.aau.dk Title: Emptio - A mobile phone application for self- service shopping in Harald Nyborg. Theme: Human Computer Interaction Project Timeframe: SW9, 1. September, 2010 - 4. Januar, 2011 Project Group: d512a Group Members: Carl Rasmus van der Burg Tórhallur Kirke Alen Rokić Supervisor: Jan Stage Abstract: Shopping is a part of our everyday life and with the current proliferation of mobile phones this report will examine whether Near Field Communication and mobile technology can be utilised to improve self-service shopping. A field study on self-service shopping is conducted with customer interviews and observations in Harald Nyborg. Furthermore, a focus group with the employees is conducted in order to understand their experiences on self-service shopping. Based on the field study, a design for a proof-of- concept mobile phone application is made to address some of the problems that exist in the current form of self-service shopping. The design consists of a sce- nario, followed by use cases, requirements and a class diagram. Although the design is not implemented in this report we examine what options we have to implement the design on contemporary mobile plat- forms. Finally, the design is reflected upon in the conclusion. Copies: 5 Total Pages: 44 Appendices: 5 Paper Finished: 4th of January 2011 The content of this paper is freely available, but publication is only permitted with explicit permission from the authors.
Transcript
Aalborg UniversityDepartment of Computer ScienceSelma Lagerlöfs Vej 3009220 AalborgTelephone:(45)96358080http://www.cs.aau.dk
Title:Emptio - A mobile phone application for self-service shopping in Harald Nyborg.
Group Members:Carl Rasmus van der BurgTórhallur KirkeAlen Rokić
Supervisor:Jan Stage
Abstract:Shopping is a part of our everyday life and with thecurrent proliferation of mobile phones this reportwill examine whether Near Field Communicationand mobile technology can be utilised to improveself-service shopping. A field study on self-serviceshopping is conducted with customer interviewsand observations in Harald Nyborg. Furthermore,a focus group with the employees is conducted inorder to understand their experiences on self-serviceshopping.
Based on the field study, a design for a proof-of-concept mobile phone application is made to addresssome of the problems that exist in the current formof self-service shopping. The design consists of a sce-nario, followed by use cases, requirements and a classdiagram. Although the design is not implementedin this report we examine what options we have toimplement the design on contemporary mobile plat-forms. Finally, the design is reflected upon in theconclusion.
Copies: 5
Total Pages: 44
Appendices: 5
Paper Finished: 4th of January 2011
The content of this paper is freely available, but publication is only permitted with explicit permission from the authors.
Signatures
Carl Rasmus van der Burg
Tórhallur Kirke
Alen Rokić
iii
Preface
This report is written by three software engineering students, currently attend-ing ninth semester at Aalborg University. The course of the project was com-menced on the 1st of September 2010 and the report was handed in on the 4thof January 2010.
The focus of the report is the Near Field Communication (NFC) technologyand how it can be utilised in the shopping context. The report concentrateson analysis of this technology and a design of an application that utilises thistechnology in a specific case. The case is conducted in collaboration with Har-ald Nyborg, in Aalborg and we would like to use this opportunity to thankeveryone from Harald Nyborg for their help and effort. We would also like tothank our supervisor Jan Stage for his advice and guidance during this semester.
The report will concentrate on subjects related to computer science and thusit is assumed that the reader has equivalent knowledge in the field of computerscience, as that of an ninth semester software student.
Two types of source references are used throughout the report. A referenceplaced after a period, refers to the given section, and if the reference is placedbefore a period or in a sentence, it refers to the particular sentence or word.Sources to the references used throughout this report can be found at the endof the report in the bibliography.
In the last decade we have seen a phenomenal proliferation of mobile phones.Most people carry one and use them frequently for various tasks. The trendshows that the growth of mobile phone subscriptions has been growing rapidly.The latest statistics from The International Telecommunication Union (October2010) shows that there will be 5.3 billion mobile subscriptions by the end of2010 [?], see figure 1.1. That is over 70 percent of the world population. Thesephones are not only used for communication. Lately there is an increasinginterest for using mobile phones for interaction with people, places and realworld objects. People can use their mobile phones to take pictures of visualmarkers such as data matrix codes. After recognising the code, the informationrelated to the code is displayed on the phone. GPS positioning is used forvarious location based services, such as "find nearest restaurants". Other areasof mobile interaction are smart objects such as advertisement posters.
Figure 1.1: Mobile subscriptions for the last decade.
1
CHAPTER 1. INTRODUCTION
One of the building blocks for this trend in ubiquitous computing and commu-nication, is the easy and natural interaction between humans and technology.Acquiring information, initiating actions, transferring data, performing finan-cial transactions, etc, can be done by touch-based interaction, which is a wayof establishing connections and exchanging information between mobile devicesjust by bringing them close to each other. It is interesting to see how technologycan be utilised in different shopping practices.
Shopping is an activity that can be done in different ways. Shoppers can pur-chase products online from various web sites and have the goods delivered totheir home. Customers can also order products from a catalogue via telephoneor mail, or order from TV or radio advertisements. The third category is whatwe call in-store shopping. This is where customers are physically in a store ora supermarket. In-store shopping can be done in various ways, which we choseto name and categorise as trolley shopping, self-service shopping, and mixedshopping. The description is as follows:
Trolley shopping This is known from grocery shopping in supermarkets, wherecustomers are browsing products in the store, selecting them from theshelves or aisles and placing them in a trolley or shopping basket. Theythen proceed to check-out where they pay for the products. The increasinginteractivity is evident today, where one of the areas in which interactivetechnologies are being introduced is in trolley shopping. Stores, such asIKEA have introduced self-checkout terminals as a supplement to the ex-isting checkout counters. Customers use the bar code scanners to scantheir items and when all the items have been successfully scanned, cus-tomers pay for their purchases with their credit card, collect their receiptand proceed to the exit. Other stores, such as Bilka and Føtex have in-stalled bar code scanners at various places in the store, where customerscan get information about the products. Both of these examples can beseen in figure 1.2. These are examples of how technology can be utilisedin a shopping context. Other examples can be seen in Appendix A.
Self-Service shopping In some stores, customers pick up a list and a pencil atthe entrance and walk around the store with this list, on which they writespecific information for a product they would like to purchase. Theseproducts are exhibited in the store, with information such as price, productnumber, size, etc. Customers write which products they want down onthe list made especially for the purpose. This is the practice used byIKEA and Harald Nyborg, and their lists can be seen in Appendix E.Since trolley shopping has evolved with new technology it is interesting tosee if new technology also can be introduced to self-service shopping.
Mixed shopping Mixed shopping consists of both Trolley shopping and Self-service shopping in the same shopping practice. Besides having to writeproducts on the list, both IKEA and Harald Nyborg have some productsin the store, which can be picked up in a shopping basket or a trolley andpaid for at the check-out.
2
Figure 1.2: IKEA’s self-checkout terminal and a bar code scanner in Føtex storein East Aalborg.
Furthermore, the proliferation of mobile phones with increasing capabilities andnew technologies makes it possible to develop applications that introduce moreinteractivity while shopping. One of these new technologies is the Near FieldCommunication (NFC). This technology enables seamless wireless communica-tion between devices such as mobiles phones and Radio Frequency Identification(RFID) tags [?].
It enables users to access information and services by simply holding the mobilephone close to objects that are equipped with NFC-compatible tags. With theuse of this technology it is not hard to imagine new scenarios in retail shopping.While shopping products in a store, the customer could use the mobile phoneto scan products, add them to a virtual shopping basket, place the order andfinally pay for the products on the mobile phone. Thus the research questioncan be summarized as follows:
• Is NFC an appropriate technology for self-service shopping?
• Are there problems in some areas of self-service shopping?
• Can we design a mobile phone application that addresses the discoveredproblems of self-service shopping?
The following provides an overview of the content in the report.
Chapter 2 describes the technology platform that will be used for the systemunder development. The focus is on NFC, what this technology is andhow it works, and some examples of its usage.
Chapter 3 looks at the shopping practice at Harald Nyborg in Aalborg andpresents the field study that was conducted. The field study consistsof customer interviews and a focus group with the employees of HaraldNyborg.
3
CHAPTER 1. INTRODUCTION
Chapter 4 presents the design for our system. The design parts consist of aclass diagram, overall use case diagram, use case specifications, state chartdiagrams, and other design related elements.
Chapter 5 is a look at how this proof-of-concept can be implemented
Chapter 6 concludes on the research questions presented in the introduction andgives a overall conclusion on the report. Finally, limitations are listed.
Enjoy your reading.
d512a
4
2Technology Platform
Over the last couple of years, mobile phones have become a popular platformfor developing diverse applications and the current trend of mobile phones canbe characterized as a move from a single purpose device to a multi functionaldevice. New technologies like Near Field Communication (NFC) extend thecapabilities of mobile phones. Invented by Philips and Sony, NFC is a noveltechnology developed especially for mobile phones. Together with Nokia, Sonyand Philips founded the NFC Forum in 2004 in order to promote short-rangewireless connectivity technology [?].
This chapter describes what NFC is, its potential and practical uses, and acomparison with other technologies. The decision to focus on NFC is based onour personal interest in this promising technology that is gaining more attentionand proliferation, which can be seen in the number of projects around the worldinvolving this technology [?]. Big companies like Google, have lately endorsedthe use NFC in mobile phones as a form of a digital wallet and see NFC usedin various applications. The CEO of Google, Eric Schmidt has lately revealedNexus S mobile phone featuring built-in NFC functionality [?].
2.1 What is NFC?
The Near Field Communication technology is a short-range radio communica-tion technology that relies on touch-based interaction [?]. NFC allows usersto establish connection and exchange information between NFC-enabled mobilephones and other handheld electronic devices, by holding them close to eachother. It enables users, to access content from another device or NFC tag atranges up to ten centimeters. In other words, touch-based interaction does notnecessarily mean devices must physically touch each other. They have to bewithin range. Thus when we write touch, it means that it represents touch-based interaction.
NFC and Radio Frequency Identification (RFID) are often used incorrectly asexchangeable terms. Both technologies use reading devices to gather data storedon passive tags that are attached to objects. NFC is based on the same technol-ogy as RFID. This makes NFC devices compatible with existing RFID systems.
5
CHAPTER 2. TECHNOLOGY PLATFORM
A quick comparison of NFC with other wireless technologies like infrared andBluetooth, can be seen in table 2.1. It should be noted, that the Range variesand is not necessarily exact with today’s standard, as these four communicationtechnologies are constantly evolving. Usability states the original goal for thetechnology and how easy it is to use. It can be seen that NFC is more humancentric, where as Bluetooth is more data centric, since it’s original goal was totransfer data between Bluetooth compatible devices. Furthermore, Bluetooth’susability is rated as Medium, since pairing of two devices is necessary in orderto establish connection, where as the other technologies do no require any setup in order to communicate. Selectivity states how one can "select the otherdevice".
RFID is rated as Partly given, meaning that it is not obvious which device isselected if there are several tags placed close to each other, and the readingdevice is, e.g., 2 meters away. NFC on the other hand reads tags at close range,and that is why it is rated as Given. By High it is meant that it is it does notrequire line of sight, and Security rating is described in section 2.4. Infraredneeds line of sight in order to establish communication, e.g., a remote for thetelevision, where as Bluetooth needs to go through the process of pairing devices.Although the table might be biased slightly, since it is from the NFC Forum andthus favoring the NFC technology, it still gives a good overview and comparisonwith other technologies on the market.
NFC RFID IrDa BluetoothSet-up time < 0.1ms <0.1ms ∼ 0.5s ∼ 6 secRange Up to 10cm Up to 3m Up to 5m Up to 30m
Usability Human centric Item centric Data centric Data centricEasy, intuitive, fast Easy Easy Medium
Selectivity High, given, secu-rity
Partly given Line of sight Who are you?
Table 2.1: Comparing NFC to other close range communication technologies(NFC Forum [?] )
2.2 Comparing different interaction techniques
The technologies in table 2.1 have different interaction techniques. NFC andRFID rely on touch-based interaction. Pointing is used in infrared and scan-ning is used with Bluetooth. It is interesting to see which of these interactiontechniques is most preferred by users.
Rukzio et al. [?] focus on the interaction between a user and a smart object inthe real world. These smart objects are products that have been augmentedwith different sensors or chips, e.g., a microwave equipped with an RFID chip.A user interacts with a mobile device, and this mobile device interacts with asmart object in the real world. The article presents an analysis and an exper-imental comparison of three interaction techniques, which they have named astouching, pointing and scanning.
6
2.3. NFC OPERATION MODES
Touching technique is where a user brings his mobile phone into close prox-imity with the smart object, when he wishes to interact with it. He is thenpresented with related information on his mobile phone. A common way toimplement this interaction technique is to use technologies like RFID and NFC,which means it is not required to physically touch the objects. It is sufficient tobring the mobile phone close to the smart object. By using the pointing inter-action technique, the user can select a smart object by pointing with a mobiledevice at it. Lastly scanning refers to the use of a wireless scanning methodto discover nearby smart objects. These object are then listed on the mobilephone, and selecting on of them results in displaying its services. The analysisshowed that the participants in their study preferred the touching technique,when they were close to objects. Especially older users preferred this interactiontechnique, since they wanted to avoid as much input on the mobile phone aspossible. Furthermore, touching was regarded as the most secure one, the mostnatural and intuitive, error resistant, and quicker than pointing and scanning.The other two require more cognitive effort.
This fits well in our a shopping context, where customers are in fact standingclose to products when they are browsing them in a store, and together withthese other attributes, natural, error-resistant etc, makes it for a good incentiveto chose this interaction technique as the main interaction for a shopping relatedapplication.
2.3 NFC operation modes
NFC has three communication modes. The first mode is called Read/Writemode, where a NFC device reads/writes data from a tag. This is similar tothe way RFID works, where a RFID reader device is active and reads a passiveRFID tag. In this mode, the NFC device acts as the initiator and the tag as atarget. The second mode is called Peer-to-Peer, where two devices communicateand exchange data. It is similar to Bluetooth, except that instead of pairingtwo devices to work together, they can simply be brought close to each otherto establish a connection. Lastly, the Card Emulation allows the NFC deviceto emulate a smart card, meaning that the NFC device itself acts as an NFCtag and appears to an external reader the same as a traditional smart card.For instance, the phone can behave as a payment card and be used to purchasegoods and services.
2.4 Examples of NFC usage
These modes give rise to different examples where the capabilities of a NFCenabled mobile phone, can be utilised:
• Smart objects: NFC tags can be attached to physical objects such thatusers can pick up information stored in those tags. For instance a tagcan store a product id, a phone number, an URL, etc. A typical use casemight be where a user touches a smart poster, that has a tag containingan URL that points to the trailer for the movie. SmartTouch was a projectin Finland, where the city Oulu had about 1500 tags installed in buses,
7
CHAPTER 2. TECHNOLOGY PLATFORM
bus stops, the theater, a restaurant, and a pub, that could be read by amobile phone. [?]
• Contactless payment: Due to the short range of communication, NFCtechnology is an ideal choice for secure transaction, since it is more immuneto eavesdropping as well as interferences. Furthermore, man-in-the-middleattacks are not possible to perform [?]. Users can use NFC enabled mobilephones as digital wallets to make payments, the same way they do withtheir credit card. This type of functionality is already common in Asia,with Japan as the leading country [?]. MasterCard and Visa are bothmembers of the NFC Forum, and both companies have been involvedin projects that utilise NFC technology for payment, e.g., MasterCard’sPayPass [?] and Visa’s payWave [?].
• Transportation: NFC can be integrated in the public transport system, sousers can pay for their tickets with their mobile phones, more commonlycalled e-ticketing. An example of this is the "Touch & Travel" projectin Germany in 2008, where Deutsche Bahn, rail operator, had a pilotprogram where travelers used their mobile phones to touch a Touchpointbefore entering the train. The passengers were billed once a month. [?]
• Health care: Patients can have NFC tags on their bed, so that the medicalstaff can quickly get information about the patient, such as what treat-ments the patient has had, what he is allergic to, etc. Elderly can useNFC-enabled devices to touch NFC tags on meal menus to easily select ameal in nursing homes. O2’s Homecare system based on NFC is a pilotproject, where home care workers can download patient records by touch-ing NFC tags installed in the patients’ home and patients can as welltouch these tags to see when the next visit from their care worker is. [?]
These are some of the examples what NFC technology can be used for. For NFCto really take off and be integrated into our everyday lives, the proliferationneeds to be more successive. The adaptation has been slow when compared toother technologies integrated into mobile phones, such as the camera.
2.5 Using technology to tag objects
The main contribution of previously mentioned smart objects comes fromWant’set al. [?] work on using RFID tags to bridge the physical and virtual worlds.It is the precursor for much of the research involving tagging objects. By aug-menting everyday objects with RFID tags, actions that are naturally associatedwith their form, could be initiated by simply scanning objects. These invisibleinterfaces as Want calls them, leverages objects of the physical world with theadvantages of computational world. For example, by augmenting a book withan RFID tag, and later scanning it with a computational device, an associatedvirtual document is displayed. It can also be linked to an associated service,such that it is linked to the corresponding web site of an online book seller andis ready to be ordered. In their prototype they demonstrate the augmentationof books, documents, business cards, wrist watches, etc., to establish links toservices.
8
2.5. USING TECHNOLOGY TO TAG OBJECTS
So what about other technologies? Surely NFC is not the only way to tagobjects. Bar codes and data matrix codes are another possibility, see figure 2.1.There exists applications for mobile phones that can read both of these codesby the use of the integrated camera. Bar codes have existed for a long timeand have been used extensively in grocery stores. They are used to label manycommercial products, however the size of these labels is constrained by thescanning technology. Tagging objects in post-process is harder with bar codesand data matrix codes than tagging them with NFC tags. It is also easier totag objects with more curved shapes, than it is with bar codes and data matrixcodes, since they require a flat surface. They also risk the danger of being cutand bar codes require holding the camera at a proper location and angle inorder to get a correct scanning. NFC tags have no such requirement and theyare more robust. They don’t degrade over time and are more resistant to dustand dirt. Furthermore NFC tags have no line of sight requirement.
Figure 2.1: Bar code and data matrix code
SummaryThis chapter has given insight into NFC and as such, we see NFC as a promisingtechnology platform to be used in a self-service shopping. The card emulationmode, makes it possible to use the mobile phone as a digital wallet for purchas-ing products, which fits well in the shopping context. Furthermore, productsin the store can be tagged with NFC compatible tags, and together with theNFC-enabled mobile phones, we can lay ground for a design of a mobile phoneapplication that utilise NFC for self-service shopping. A scenario describing thisusage will be given in a later chapter.
9
3Field Study
This chapter gives an overview of the case and, furthermore, describes the fieldstudies conducted. The field study of Harald Nyborg (HN) consists of a focusgroup and a set of customer interviews. Some of the reasoning in this chapteris inspired by Research Methods [?].
3.1 Case overview
For this project we have made arrangements with HN in Aalborg in order touse them as a case. The HN store is a department store that deals with a widerange of products such as tools, CD-players, kids toys and bicycles, see figure 3.1.The shopping practice in HN is mixed shopping, however, it is mainly self-serviceshopping which fits well for our purpose to use them as a case.
Figure 3.1: A Harald Nyborg store.
A sketch of how the inside of the store looks can be seen in figure 3.2. Therectangular boxes marked with an ’S’ are shelves with products. Each shelfhas an unique number in order to identify it in the store. The ’L’ in the boxnear entrance is where the shopping lists for customers are placed and the boxmarked with a ’C’ is the counter.
3.2. CUSTOMER INTERVIEWS AND OBSERVATIONS
Figure 3.2: A sketch of the inside of a Harald Nyborg store.
Shopping processThe way customers shop in HN is that when they enter the store there is astand with some shopping lists for the customers. The customers can take oneof those shopping lists and use it when browsing the store. All the products inthe store have an identification number that the costumers can note down ontheir shopping list along with the price of the product and the quantity theywant. It is also possible to write down the name of the product. When thecustomer is done browsing she will go to the counter and hand in the shoppinglist to the one of the store clerks, who then will enter the information on a com-puter and then give the customer an order number. The storeroom staff thenreceive a printed list of the order and they then fetch the products on the listand return them to the store counter via a conveyor belt. Then one of the storeclerks will check the order number of the returned order and place the ordernumber on a big screen so the costumers know their order is ready for pick up.
We wanted to understand the shopping process at Harald Nyborg and the cus-tomers’ experience in their natural habitat before designing the system andtherefore we opted to have a direct discussion with concerned participants, whichin our case were customers and employees. We chose to conduct interviews withthe customers and convene a focus group with the employees.
3.2 Customer Interviews and Observations
The interviews were designed as semi-structured interviews. When doing aninterview there can be a mismatch between what the interviewees say and whatthey actually do and therefore we combined the interviews with an observationto be able to filter out some of those situations. This method was chosen, inorder to get in the field and observe pros and cons of the current shopping systemused at Harald Nyborg, along with getting the customers’ insights using a Think-Aloud protocol and interview questionnaires. The interview questionnaire canbe seen in Appendix B. Picture 3.3 shows an interview in action. The picturecontains two group members, one interviewer, one note-taker, and the customer,which can be seen in the background in blue overalls.
11
CHAPTER 3. FIELD STUDY
Figure 3.3: Observation of a customer in Harald Nyborg.
The goals for the interviews were as follows:
• At least three customers
• Preferably one young, one middle aged, and one old customer
• Preferably at least one of each gender
• At least one novice and one expert customer
We chose to do interviews because they provide the ability to go deep and getdetailed responses from participants, but they also require conversational andobservational skills, such as understanding non-verbal cues. These skills requirepractice and experience to develop. Another incentive for conducting interviewswith the customers is, that we get closer to reality when the customers are inthe store and are shopping.
Setting
The interviews took place in a Harald Nyborg store located in Skalborg, Aalborg.We set up close to the entrance in order to be visible to the customers.
Participants
All group members participated in the interviews. There were different rolesduring each interview; interviewer, audio recorder and note-taker. The note-taker was also responsible for taking pictures. The roles were switched in someinterviews. Seven interviews were conducted with customers where two of thecustomers had little experience in shopping at Harald Nyborg and five customerswith medium to high experience. The age span was 22-82 years, and the meanage was 46 years, SD = 19.49.
12
3.2. CUSTOMER INTERVIEWS AND OBSERVATIONS
Materials
In order to gather data from the customers, paper sheets were used to writeinformation about the customers and notes from the interviewer. A digitalcamera was used to take pictures and an audio recorder was used to recordthe audio during the interviews. Coffee and biscuits were also offered to thecustomers. All group members wore identical t-shirts that reflected that wecame from Aalborg University to avoid any misinterpretations.
Procedure
The customer interview session was started by a small meeting with the storemanager to let him know that we had arrived and to get his approval of settingup. Then we positioned ourselves close to the store entrance where coffee andbiscuits were placed. We then proceeded to intercept incoming customers andask if we could ask them a few questions. If the customers agreed then we wouldfollow the customer around, observing him during shopping, taking notes andasking questions. After the interviews were completed we spent some time inthe store observing customers and their shopping habits.
Data Collection
After the interviews were finished the data collected consisted of audio record-ings of each interview, pictures of some of them, notes and information fromeach interview.
Data Analysis
The data collected was analysed by transcribing each interview and then analysingthe transcription. The transcriptions were used to generate ideas for the prod-uct in question. Additionally, the notes and customer information was used foridea generation. These ideas were used to generate features for the system un-der development, discover possible user interactions problems and benefits and,finally, for writing use cases for the system.
3.2.1 ResultsThe yield of the customer interviews was primarily ideas for the product. Wefound out that the customers can be categorised; novice customers and expertcustomers. The novice customers can again be split into 2 sub categories. Someknew the current system at the store and others did not. The expert customerswere customers that had great experience of using the current system. Thegoals that were set for the interviews were all fulfilled. We managed to get seveninterviews with customers of various ages and gender. Of the seven people therewere two novice customers and five expert customers.
13
CHAPTER 3. FIELD STUDY
These are the results we got from the interviews and observations:
Problems Novices (2) Experts (5) Observations1. The shopping list was hardhard to find
X X
2. Customers did not knowwhat the shopping list was for
X
3. Products were difficult tofind
X X
4. Shopping list was preparedfrom home
X
5. Most customers were men X6. Customers browsed a lot X7. Customers waited approx-imately 10 minutes while theorders were being processed
X
8. The average age was high X9. Customers did not havesmartphones
X X
Table 3.1: Results of customer interviews.
1. Both novice customers experienced the problem.
2. Only one of the novice costumers did not know what the shopping list wasfor.
3. Both novice costumers had problems with finding the products they werelooking for. Furthermore, we observed several customers having problemsfinding products.
4. Three of the five expert costumers had a shopping list prepared from homeand the two other knew what products they wanted as well as where theywere located.
5. We observed that the majority of customers in the store, at all times, weremen.
6. We observed that several customers were browsing the store without buyingany products
9. None of the interviewees had smartphones and one of the expert customersdid not have a mobile phone.
As seen in table 3.1, the results are divided into novices, experts and observa-tions. The novices and experts results are the results we got from the transcrip-tions of the interviews, while the observations were what we observed in generalat the store as well as while doing the interviews.
14
3.3. FOCUS GROUP
3.3 Focus group
Before starting the idea generation for the project, we wanted to have a focusgroup together with the employees at Harald Nyborg. The focus group was setup together with our contact person at Harald Nyborg. The focus group wasneeded in order for us to understand the needs, frustrations and concerns theemployees might have and what they think the customers might have. Also toask them, what they would like to able to do, regardless of the limitations ofthe current technology. Before the focus group, we made an agenda, that canbe seen in Appendix C, which contained a list of questions.The goals of the focus group were as follows:
• Focus group should consist of at least four employees.
• Focus group should consist of employees with experience from the shopcounter and employees with experience from stockroom.
• Discover possible problems with the current system.
• To get the employees to generate ideas.
Setting
The focus group took place at the employee canteen in Harald Nyborg, seefigure 3.4 The date and time was found through our contact person, which alsois the manager of Harald Nyborg in Aalborg. The focus group was held withone person recording audio and taking notes and one interviewer.
Figure 3.4: Focus group with employees of Harald Nyborg.
Participants
The focus group consisted of 2 members of our project group and 10 peoplefrom Harald Nyborg, including our contact person. We were lucky to be ableto get every person in the store as the focus group was held on a Friday, which
15
CHAPTER 3. FIELD STUDY
is when most employees are at work, before opening time. However this meantthat the focus group was held during their morning break and we only had 25minutes to complete it. This, in turn, meant that people were relaxing andeating breakfast during the focus group and thus might not be very focused onthe focus group as they could have been.
Materials
During the focus group an audio recorder, a digital camera as well as smallquestionnaires were used to gather data. Furthermore, an agenda was used tokeep the focus group on track.
Procedure
The focus group was handled with one group member being the focus groupinterviewer and the other member recording the audio, taking notes and finallytaking pictures of the event. First an introduction to our project was given tohelp the participants understand what our aim was. Then a brief explanationof how the focus group was going to be held was given. Finally, the actual focusgroup event started by going through our agenda and asking questions. Thefocus group was held semi- structured which meant that some questions wereasked which were thought of during the focus group. The entire focus groupwas audio recorded and the transcription can be seen in Appendix D.
Data Collection
After the focus group was held we had gathered the data we wanted and thisdata consisted of audio recording of the entire focus group and some notes anda few pictures. Most of the employees were very quiet and hard to get to speakand only two people were talkative. Although most were very quiet we gotresponses from all employees at some point.
Data Analysis
Shortly after the focus group was held, a brief resume was written accordingto our memory of the event. The audio recording was transcribed to ease theanalysis of the data. The transcription was analysed and key elements wereextracted from the text. These key elements consist of data useful for generat-ing features for the system. Information such as the problems they have withinterpreting the customers’ handwriting, mistypings and such was importantdata we extracted. We also learned that the customers’ average age is quitehigh, which also means that the system we want to develop should take thatinto account. We also learned that many customers have problems using theshopping list and although this might be related to the customers’ average agewe still need to take this into account as this might imply that Harald Nyborgcustomers’ can have problems learning how to adopt new technology. Anotherthing we learned was that the different views of the mobile application shouldnot contain too much data as this might confuse the customers.
16
3.4. REFLECTION ON OUR FIELD STUDY
3.3.1 ResultsThe goals for the focus group were all fulfilled. We wanted at least 4 employeeswith both experience from the shop counter and the storeroom. We got 10people with experience from all areas of Harald Nyborg. We succeeded in gettingthe employees in generating ideas for the future system, although they did notgenerate as many ideas as we had hoped for.
Problems experienced By employees By customersShopping list was hard to find XPurpose of shopping list was sometimes unknown XMistyping the product number X XUnreadable handwriting on the shopping list XMost customers were men XProducts were hard to find XAverage customer age was high X X
Table 3.2: Problems experienced by employees.
Table 3.2 is divided into customers and employees, where the customers sec-tion is problems existing on the customer side of the shopping process and theemployees section is of course problems existing on the employee side. Further-more, The store manager at Harald Nyborg stated that they are trying to makethe store more attractive for a younger audience.
3.4 Reflection on our field study
We had set goals for both the focus group and the interviews and all of thesegoals were reached. The interviews helped us to understand customers needsand find out if there were any frustrations with current practices. The questionsasked were high-level questions so we could find out if there was something thecustomers would like to do, that they currently are not able to do. Before de-ciding to conduct interviews, it was obvious to us, that novice customers mighthave different perspective from expert customers. That is the reason we set thegoal to include representatives from both groups, since it would provide a morecomplete picture of the situation.
The decision to make our interviews semi-structured instead of fully-structuredis because, it enables us to dig deeper into interviewees comments and thusachieve both breath and depth. These comments can also lead to questions wehave not thought of or overlooked. This flexibility is harder to achieve withfully-structured interviews which follow a rigid script.
One of the weaknesses with our interviews was that we only had a low numberof interviewees. With more interviewees we could have gotten a better overviewof the severity of the found problems and discovered other problems as well.Another problem was that the interviewers were inexperienced in interviewingand therefore it was difficult to be consistent in flow and performance in theinterviews.
17
CHAPTER 3. FIELD STUDY
For the focus group, the main yield was the problems that the staff at HN haveexperienced and what they have noticed that the customers have experienced.Focus group suffered from silence and it was hard to get some employees totalk. Interpreting silence is hard, since it can depend on many factors. It canbe that the interest in the topic is lacking. There is also the possibility thatthe participants can be intimidated by their colleagues and thus it is harderto criticise the current system if your boss is sitting next to you. Participantsmight simply just be bored or even tired. It can also be that our timing wasnot the best, since it was held during the breakfast, or maybe the focus groupwas to large.
It is hard to say if we could have gotten a more in-depth conversation if therewere fewer participants. Maybe we should have convened the focus groups withfive participants and held several sessions as Kruger recommends [?, Chp. 8,p. 192]. However, the results we got from the focus group session were morevaluable than the results from the interviews, since it raised more ideas for thesystem and highlighted issues we did not think of, such as employees mistypingproduct numbers. All in all we think that the focus group session was moresuccessful than the interviews. Even though the data from the session hadlower quantity than the interviews, the quality of the data was better.
18
4System Design
Based on the results and insights from the previous chapter, the design choicesfor the system under development are described in this chapter. We have chosenEmptio as the name for the system under development. This inspiration comesfrom Latin, where the word emptio means; buying or a purchase. A descriptionon each of the elements used in designing Emptio, is given. These elementsconsist of a class diagram, sequence diagram, an overall use diagram, use casespecifications, and state chart diagrams. Additionally, this chapter describes thepurchasing process with Emptio compared to the current purchasing process.
4.1 Purchasing Scenario
This section describes a scenario of a customer that uses Emptio to shop in aHarald Nyborg (HN) store. The scenario is as follows:
A customer enters a HN store and starts browsing around thestore. The customer wants to find and read the description of a prod-uct he is interested in. He holds his NFC-enabled mobile phone closeto a product’s tag and the information, e.g., a picture and some de-scriptive text, for that product is then displayed on his mobile phone.He presses on the button for purchase, and this adds the product tothe virtual order list on his mobile phone. Later on, the customerwants to find the physical location of a product in the store. This isachieved by searching for a product in the catalogue on the mobilephone and selecting a product. After finding all desired products thecustomer chooses to checkout on the application. Upon completingthe payment, he is given a queue number. When his queue numberis displayed on the screen at the counter, the customer can pick upthe items and the purchase is completed.
This scenario was used to illustrate a typical purchase with Emptio, and it canbe seen that it resembles the current purchasing process. It just happens ona mobile phone instead of the customer having to manually write the productnumbers down on to a piece of paper. An illustration of the purchasing processcan be seen in figure 4.1.
19
CHAPTER 4. SYSTEM DESIGN
Emptio Customer
Send order
Forward order
Forward payment
credentials
Finish order
HN
Validate credentials
Pick up products
Send payment info
Accept or reject
payment
Validate order
Figure 4.1: The purchasing process.
4.2 User Requirements
This section describes the use cases and the user requirements of Emptio.
4.2.1 Use Cases
Use cases are used to describe a system’s behavior as it responds from requestsfrom different stakeholders. The primary actor is the stakeholder who or whichinitiates an interaction with the system in order to achieve a goal. An actor is aperson or another system with a behavior. The system responds to the requestsand helps to fulfill the primary actor’s goal. A team can use them as a com-munication device to discuss the upcoming system. Use cases can be used todescribe business processes but often they are used to document the behavioralrequirements for an application. They simply tell how the system will behavein use.
20
4.2. USER REQUIREMENTS
The use case form used in this report is adopted from Alistair Cockburn andmodified slightly. First of all a use case has a name and a number, so it is easierto refer to them. A primary actor is the one having a goal to accomplish, andwhich the system will support. The scope identifies the system that we are dis-cussing about, and level describes the level that the use case is written to andit can be on of these; summary, user goal, or sub-function. The preconditionsis what must be satisfied before the use case runs. The trigger is an event thatstarts the use case. The description part contains the main scenario in which anactor goes though some action steps and achieves a result where nothing goeswrong. The extensions part describes what can fail during different steps in themain scenario. The sub-variation part shows alternative steps or flow an actorcan take to accomplish his goal.
The overall use case diagram is presented in UML in figure 4.2. The primaryactor is the customer in all four use cases depicted in the figure. The followingpages contain detailed use case specifications for all the four use cases, anddescribes the behaviour of the system as the primary actor interacts with it.The use cases specifications are supplemented with state chart diagrams for avisual illustration.
Figure 4.2: Overall use case diagram
21
CHAPTER 4. SYSTEM DESIGN
Touch the initialisation tag
The state chart diagram for the first use case from figure 4.2 is depicted infigure 4.3. This use case describes what happens when a customer touches theinitialisation tag. A detailed description can be seen in table 4.1.
Touch the initialisation tag
Application URL displayed
Application installed
Install application from URL
Figure 4.3: State chart diagram for Touch the initialisation tag.
22
4.2. USER REQUIREMENTS
Use Case: 1 Touch the initialisation tag.Goal in context Use the initialisation tag.Scope & Level Emptio, User GoalPreconditions Customer is in the store and has a mobile
phone.Success End Condition The customer has touched the initialisation
tag and the application is downloaded and in-stalled on the mobile phone.
Failed End Condition The download and installation of the applica-tion has failed.
Primary, SecondaryActors
Customer
Trigger Customer touched the initialisation tag withan NFC-enabled mobile phone.
Description Step Action1 Customer’s mobile phone opens a URL
received from the initialisation tag.2 Customer installs the application on the
mobile phone.Extensions Step Branching Action
1a Initialisation tag contains the wrongURL.
1b The URL points to the correct link butthere is no connection to the server.
2 Customer is unable to install the appli-cation.
Performance Touching the initialisation tag and receivingthe URL should not take more than 3 seconds.
Table 4.1: Touch the initialisation tag.
23
CHAPTER 4. SYSTEM DESIGN
Read product details
The state chart diagram for the second use case from figure 4.2 is depicted infigure 4.4. This use case describes what happens when a customer wants to seethe information for a product. A detailed description can be seen in table 4.2.
Product's info displayed
Touch a product’s tag
Press ”mere info” button
Additional info displayed
Figure 4.4: State chart diagram for Read product details.
24
4.2. USER REQUIREMENTS
Use Case: 2 Read product detailsGoal in context Find and read a product’s description.Scope & Level Emptio, User GoalPreconditions Emptio is running and a connection is estab-
lished to HN’s system.Success End Condition The customer has read the desired product’s
description.Failed End Condition The product’s description can not be read.Primary, SecondaryActors
Customer
Trigger Customer holds an NFC-enabled mobilephone close to a product’s tag.
Description Step Action1 Upon identifying a product, the cus-
tomer is presented with a view thatshows basic information for that prod-uct.
2 Customer presses the "More info" but-ton to reveal more details about theproduct.
3 The details are then displayed underthe basic description.
4 Customer can now read this extra in-formation.
Extensions Step Branching Action1a Customer is not aware of where the
product’s tag is.1b Customer is not able to get a successful
reading of a product’s tag.2a Customer can not see the "More info"
button.2b Pressing the "More info" button does
not reveal extra info.4a The content is not presented clearly and
is cluttered.Subvariations Branching Action
1 Customer can also select a productfrom the catalog.
PerformanceIdentifying a product and displaying the infor-mation for that product should not take morethan 3 seconds.
Open IssuesHow to provide the option to compare similarproducts?How to generate a list of related products?
Table 4.2: Read product details.
25
CHAPTER 4. SYSTEM DESIGN
Find the location of a product in the store
The third use case from figure 4.2 describes what happens when a customerwants to find the physical location in the store for a product he is interestedin. This can be seen in the state chart diagram in figure 4.5, and a detaileddescription is in table 4.3.
Result set shown
Press ”Søg” button
Search field shown
Enter search term
Basic info view shown
Press on a product
Read location
System search initiated
No results found
Results foundCancel
Cancel
Figure 4.5: State chart diagram for Find the location of a product inthe store.
26
4.2. USER REQUIREMENTS
Use Case: 3 Find the location of a product in thestore.
Goal in context Help the customer find the location of theproduct in the store.
Scope & Level Emptio, User GoalPreconditions Emptio is running and a connection is estab-
lished to HN’s system. Emptio has a catalogueof the products.
Success End Condition The product’s location is displayed on the mo-bile phone.
Failed End Condition The location of the product is not shown tothe customer.
Primary, SecondaryActors
Customer
Trigger Customer wants to find a product.Description Step Action
1 Customer presses the "Søg" button.2 The customer enters the search word.3 Emptio searches in its catalogue and
displays a list of results.4 The customer selects a product from
the result list.5 The location of the product is shown to
the customer.Extensions Step Branching Action
1a Customer can not find the "Søg" but-ton.
2a The customer might misspell the searchword.
3a Emptio might return an empty resultset.
3b Emptio might falsely return a result setwithout the wanted product.
4a The customer might choose the wrongproduct.
5a The location of the product might beincorrect.
Subvariations Branching Action1 Customer can also find a product’s lo-
cation by browsing the catalogue.Performance Searching should not take more than 5 sec-
onds.Open Issues What if the location is outdated?
Table 4.3: Find the location of a product in the store.
27
CHAPTER 4. SYSTEM DESIGN
Add a product to the virtual order list
For the forth use case from figure 4.2 the state chart diagram in figure 4.6 showsthe interaction when a customer decides to add a product to his virtual orderlist on the mobile phone. Description for this use case is in table 4.4.
Product added to virtual order list
Touch a product’s tag using a NFC-enabled mobile phone
Product's info displayed
Press ”Køb” button
Figure 4.6: State chart diagram for Add a product to the virtual orderlist.
28
4.2. USER REQUIREMENTS
Use Case: 4 Add a product to the virtual order listGoal in context A customer wants to add some products to the
virtual order list.Scope & Level Emptio, User GoalPreconditions Emptio is running and a connection is estab-
lished to HN’s system.Success End Condition The product is in the virtual order list.Failed End Condition The product can not be added to the virtual
order list.Primary, SecondaryActors
Customer
Trigger Customer touches the product’s tag with amobile phone
Description Step Action1 Upon touching the product’s tag, the
customer is presented with a view thatshows basic information for that prod-uct.
2 Customer presses the "Køb" button.3 The chosen product is than added to
the virtual order list on the mobilephone.
4 Customer is presented with a view thatshows the content of the virtual orderlist.
Extensions Step Branching Action1a Customer is not aware of where the
product’s tag is.1b Customer is not able to get a successful
reading of a product’s tag.2a The "Køb" button is not easy to spot.3a No feedback provided when adding a
product to the virtual order list.4a The content of the virtual order list
does not match what customer hasadded.
Subvariations Branching Action1 Customer can also add a product to the
virtual order list by manually enteringa product id.
2 When browsing the product catalog,customer can also add a product to thevirtual order list by selecting a productin the catalog.
Performance Identifying a product and displaying the infor-mation for that product should not take morethan 3 seconds.
Open Issues How to provide the option to change the quan-tity of products?
Table 4.4: Add a product to the virtual order list. 29
CHAPTER 4. SYSTEM DESIGN
Place order and pay
The state diagram 4.7 for the last use case from figure 4.2 describes what hap-pens when a customer decides to checkout. A more detailed description is intable 4.5.
Payment info sent to HN
Customer places order
Order received at HN
Customer touches point-of-sale device
Payment verified
Queue number sent to Emptio from HN Payment rejected
When order is ready, customer picks up the order
HN verifies order and returns amount
Amount received by Emptio
Figure 4.7: State chart diagram for Place order and pay.
30
4.2. USER REQUIREMENTS
Use Case: 5 Place order and payGoal in context A customer wants to place an order and pay.Scope & Level Emptio, User GoalPreconditions Emptio is running and the connection to the
HN’s system is established. The customer’svirtual order list is not empty.
Success End Condition Customer’s order got placed and payment wassuccessful.
Failed End Condition Customer’s order did not get placed or pay-ment failed.
Primary, SecondaryActors
Customer
Trigger Customer decides to check-out.Description Step Action
1 Customer sends the order from Emptioto HN’s system.
2 HN’s system receives order, calculatesamount, and finally sends the amountinfo to Emptio.
3 Emptio receives the amount and cus-tomer pays.
4 Payment is verified in HN’s system.5 A queue number is generated in HN’s
system and sent to Emptio.6 The number in the queue is shown on
a screen behind the counter when theorder is ready for pickup.
7 Customer walks to the counter andpicks up his order.
Extensions Step Branching Action1a Customer tries to send an empty virtual
order list.2a Order is not received in HN’s system or
not all data is received.3a The amount is not received by Emptio.4a The payment is not successful.5a The queue number is not received by
Emptio.Performance N/AOpen Issues How do we ensure, that all data is sent? Can
credit card emulation be implemented prop-erly?
Table 4.5: Place order and pay.
31
CHAPTER 4. SYSTEM DESIGN
4.2.2 FeaturesThis section covers the features set for the system. The features are divided intoCore and Additional. The Core features are the minimal features the systemmust have in order to function. The Additional features are not vital for thesystem, but make it more enjoyable and generally improve the system. Thefeature categories also function as a priority as the Core features should beadded to the system before the Additional features.The features stated in this section are the results of the previous chapter.
Core features
• Touch a product in the store to see basic info for a product.
• add product to the virtual order list.
• remove product from the virtual order list.
• change quantity of a product in the virtual order list.
• place order and pay.
Additional features
• a stock status for a product.
• if a product has additional info available, such as power consumption, size,etc., the customer should be able to see it.
• if a product has some related products, they should be displayed (e.g.buying a wall socket, do you have cable, screws etc.).
• a catalogue of the products should be available on the mobile phone, sothe user can search for a product by entering e.g. a product’s name, orjust flick through the pages.
• add products from the catalogue to a virtual order list on the mobile phonebefore coming to store.
• add products by manually entering the product number.
• show queue number when order has been placed.
• a history of purchases.
• provide a help window that explains how the application is used
32
4.3. TECHNICAL REQUIREMENTS
4.3 Technical Requirements
This section explains the technical requirements for Emptio, which includes theclass diagram and quality goals.
4.3.1 Overview of EmptioThis section gives an overview of Emptio and external systems it interacts with.Figure 4.8 shows overview of the organisation, including Emptio, Harald Ny-borg’s internal system and the payment system it interacts with.
Figure 4.8: Overview of the organisation. Emptio is shaded, meaning that it isthe system under development.
Emptio is used directly by the customers. From Emptio there is an interactionwith HN’s system, where the arrow means, that customers sends informationsuch as payment credentials and the contents of the virtual order list to HN’sinternal system. From HN’s internal system, Emptio receives information suchas payment status, a queue number, and stock availability. Lastly, HN sendspayment information to the payment system, which then sends a payment statusback to HN. In order to provide an NFC-based service, HN’s internal systemneeds to be expanded to provide some back-end services, which it currently doesnot have. However, since this project is a proof-of-concept, we will not designand implement a back-end system, but rather simulate its services. Furthermore,we do not have access to Harald Nyborg’s internal system and are thus notallowed to extend it.
4.3.2 Class DiagramThis section gives an overview of the class diagram that will be used in theapplication, see figure 4.9. A description of the classes and their relations isgiven.
First of all we have made a choice to have a catalogue that contains all productsin the store. This class will be a singleton, and thus it deploys the Singletonpattern [?], since only one catalogue object will exist at run-time. Furthermore,this catalogue will have a number of different categories. This class will be re-sponsible for searching products, and so it is convenient that only one objectis performing the search of all the products in the store. In order to be upto date, there will be a method to update the catalogue to the latest revision.The Category class will represent categories like Tools, House and Garden,Dishwashers, Housekeeping, etc., and will contain all the products of a certaincategory. For instance, a hammer will be under the Tools category. Due to this
34
4.3. TECHNICAL REQUIREMENTS
composition, we have made use of the Hierarchy pattern [?], which is used toorganise objects in different levels. At the top level the Catalogue class has aCategory, which in turn has a Product.
The Purchase class and its sub parts are used to model a purchase. A purchaseis modeled as first being a virtual order list, then it becomes an order. Wehave chosen to use the Sequential Role Pattern [?], in order to indicate that apurchase goes through several roles. In other words, a purchase can only bein one role; either in the role of a VirtualOrderList or an Order. The statechart diagram for the Purchase class describes its behaviour and can be seen infigure 4.10. A VirtualOrderList object gets created when a product is addedto it. When the customer’s payment succeeds an Order object is created. Theinformation contained in the object can then be stored on the mobile phonefor future reference by the customer. When the purchase is completed thePurchase and its sub parts; VirtualOrderList and Order objects are closed.The customer can cancel the purchase any time.
Products AddedAdd Product
Add Product
Cancel Purchase
Products Ordered
Products Picked Up
Order
Payment
Purchase Completed
Figure 4.10: State chart diagram for the Purchase class.
Catalogue class: The Catalogue class contains all the products on the store.It has a categorylist which contains all the Category objects and it has a Searchmethod which is used to return a set of Product objects which contain the searchterm. The search method uses the list of Category objects to find the Productobjects.
35
CHAPTER 4. SYSTEM DESIGN
VirtualOrderList class: The VirtualOrderList class represents the vir-tual virtual order list and is used to enable the customer to add products tothe virtual order list in the mobile phone. It contains a list of ShoppingItemobjects, which contain an item id, quantity and a Product object. Also theVirtualOrderList class contains some methods to maintain the virtual orderlist, i.e., to remove, add, modify and empty the virtual order list.
Order class: The Order class is used to represent the finalised purchase. Itcontains all relevant information available after a purchase, i.e., an order date,any possible notes, a paid flag and a list of ShoppingItem objects. This is thenstored persistently for future references.
4.3.3 Quality Goals
This section describes the quality goals for the system under development. Thequality goals can be seen in table 4.6. The quality goals used in this report canbe further studied in the OOA&D book[?, s. 174].
Criterion Very important Important Less important Irrelevant Easily fulfilled
Usable: The system’s adaptability to the organisational, work-related, andtechnical contexts. It should be intuitive to use for people who know how touse smartphones, and thus it is rated Important. We aim to make it as easy touse as the current shopping list.
Secure: The precautions against unauthorized access to data and facilities.Since the system will handle payments and sensitive data from the customers ithas been rated Very Important.
Efficient: The economical exploitation of the technical platform’s facilities.Efficiency has been rated Less Important as the application will not require anymajor computations or send or receive large amounts of data.
36
4.3. TECHNICAL REQUIREMENTS
Correct: The fulfillment of requirements. Correctness has been rated LessImportant since we have a limited development time and we have chosen tofocus more on making the application usable and reliable.
Reliable: The fulfillment of the required precision in function execution. Theprecision of the function execution is deemed Very important in order to be ableto fully test the system, e.g., when a customer touches a tag the correct productshould be shown on the mobile phone. Total price must be calculated correctlyand discounts should be handled correctly.
Maintainable: The cost of locating and fixing system defects. Maintainabilityhas been deemed Important because it is intertwined with the testable criteria,which we have set to important. Therefore, we strive to get low coupling.
Testable: The cost of ensuring that the deployed system performs its intendedfunctions. Since we focus on reliability, making the system testable is deemedImportant. In order to be sure that the system is reliable, we need to test thatit is not inreliable.
Flexible: The cost of modifying the deployed system. This is rated as Lessimportant since we are focusing on making the system only for Harald Nyborgand not any other stores.
Comprehensible: The effort needed to obtain a coherent understanding ofthe system. Comprehensibility is not to be neglected, however it is still ratedLess important as we are only three developers on the system and it will not betransferred to another development team.
Reusable: The potential for using system parts in other related systems. Reusabil-ity is considered Irrelevant as this system will only be developed for one platformat a time. There is no standard for targeting multiple platforms at a time.
Portable: The cost of moving the system to another technical platform. Atthe current state, there are no plans of making the system available to othertechnical platforms and portability has therefore been rated Irrelevant.
Interoperable: The cost of coupling the system to other systems. As men-tioned earlier, this system is to be used with the existing system at HN and it isnot meant to work with other systems and therefore interoperability has beenrated Irrelevant.
37
5Implementation
In this chapter we give our thoughts on how we see this proof-of-concept imple-mented on a mobile platform. During this chapter we look at the current stateof mobile phones and their capabilities and which development tools should beused for the development.
5.1 Platform
In the consideration of platform for mobile development there are several majorplayers to take into consideration: iOS, Windows Phone 7, Android and Sym-bian OS. Additionally, there exists a framework that targets multiple platformscalled ELIPS [?]. This framework can be used to generate native code for Win-dows Phone/Mobile, Symbian OS, Android, and iOS amongst others. ELIPSsupports features such as making voice calls, accessing the mobile phone’s ad-dress book, sending SMS and taking pictures with the camera, however it doesnot currently support NFC and therefore is not applicable. However, if ELIPSwere to support NFC in the future it would be a wise choice of developmentframework because it provides a single toolkit that targets multiple contempo-rary platforms.
Because our application is a proof-of-concept and we have rated portability asirrelevant in the quality goals section, we have chosen only to develop for asingle platform. This means that we can focus on implementing more func-tionality on one platform instead of implementing less functionality on multipleplatforms. To our knowledge, Android is the only platform on which NFC hasbeen announced to be natively supported, from version 2.3. Other platformswill probably follow soon, however, since it has not been announced yet, we havechosen to develop for Android. Additionally, the first NFC enabled smartphonesthat have already been announced all run on Android, e.g., the Google NexusS (Android 2.3).
5.2. DEVELOPMENT FOR ANDROID
5.2 Development for Android
The recommended way of developing applications for Android is to use theEclipse IDE [?] with the Android Development Tools (ADT) plugin installed.This plugin adds powerful features related to Android development to the al-ready powerful IDE Eclipse. The ADT adds a New Android Project to theWizard which helps the developer to start up at new project at it practicallycreates a new project ready to be compiled to an application which can be runon Android straight away.
Furthermore, the ADT adds the possibility to emulate real world devices, suchas a mobile phone running Android OS. This gives us the opportunity to test theapplication while developing and thus making the development more efficient.
Choice of Programming LanguageThe Android platform offers the developer a lot of languages to develop appli-cations in. These include Java, C#, C++, Ruby and many others. However,Java is the most popular programming language for Android, partly becauseit is an older, yet powerful language with many existing libraries and is alsothe best supported language for Android. Based on this we will develop theapplication using the Java language. This will give us the opportunity to findmany resources such as code samples and documentation.
39
6Conclusion
First, we conclude on the three research questions raised in the introductionand present the solutions for the problems found. Afterwards, the entire reportis concluded upon and, finally, the limitations of the project are presented anddiscussed.
The first research question from the introduction chapter was:
Is NFC an appropriate technology for self-service shop-ping?
In chapter 2 on page 5 NFC was compared to other technologies, such as RFID,Bluetooth, Infrared, bar codes and data matrices. In this comparison we foundsome strengths and weaknesses of NFC:
Strengths
• It is easy to use because no setup is required.
• It can be used as digital wallet.
• It is not possible to perform man-in-the-middle attacks.
• It is hard to eavesdrop.
• NFC interaction is similar to bar codes and data matrices.
One of the most important things for NFC to be an appropriate technology isthat it can be used as a digital wallet, which enables a user to pay with thesame device used for shopping, making the mobile phone an all-in-one device,which means that the whole shopping process can be handled from the mobilephone.
NFC interaction is done by touching a product with a NFC enabled device andthis interaction is very similar to scanning a bar code with a bar code reader.
Weaknesses
• Although it is difficult to eavesdrop, it is not possible to guarantee aneavesdrop-free communication.
• NFC does not offer encryption possibilities.
• NFC has not seen high proliferation in this part of the world.
Although it is hard to eavesdrop it is not possible to guarantee that data cannot be intercepted. However, the eavesdropper has to get within the range ofthe NFC chip, which is low, in order to eavesdrop.
The NFC chip does not offer any form of encryption of the data. However thedata can be encrypted by the device that contains the NFC chip before the NFCchip sends the data.
The second research question from the introduction chapter was:
Are there problems in some areas of self-service shop-ping
In order to find problems in the existing purchasing process, we conducted afield study, as seen in 3 on page 10, which consisted of customer interviews andan employee focus group. The problems which were found are listed below.The number of participants was seven (n = 7) and the number in parenthesisfor each item below shows how many of the seven participants experienced theproblem.
• Some of the customers found the shopping list hard to find. (2)
• Similarly, some customers found the products hard to locate. (2)
• The purpose of the shopping list was unclear to some customers. (1)
• Some customers prepared the shopping list from home. (3)
• Some employees experienced that customers had mistyped product num-bers.
• Employees experienced that they occasionally mistyped product numbersinto the system.
• Employees could not always read the customers handwriting on the shop-ping list.
• Customers average age was high.
• Most customers were men.
• Customers browsed the store a lot.
• Customers waited approximately 10 minutes while the orders were beingprocessed.
• Customers did not have smartphones. (7)
41
CHAPTER 6. CONCLUSION
The third research question from the introduction chapter was:
Can we design a mobile phone application that addressesthe discovered problems of self-service shopping?
During the design phase of the project we have tried to address the problemsfound in the previous research question. Although many problems were ad-dressed and we found possible solutions for them, some of the problems cannot be solved with NFC and mobile technologies or were out of scope. Theseproblems are as follows:
• Customers average age was high.
• Most customers were men.
• Customers did not have smartphones.
• Customers browsed the store a lot.
The problems that were addressed successfully are as follows:
Some of the customers found the shopping list hard to find This problem is auto-matically solved with a mobile application as the shopping list always isin the customer’s hand, given that the customer carries the mobile phone.
Some customers found the products hard to locate By using the virtual catalogue,the customers are able to lookup the location of the products in the store.An alternative solution would be to use indoor positioning of products.
The purpose of the shopping list was unclear to some customers This problem isalso automatically solved as with the mobile phone, the physical shoppinglist is not required anymore. However, this might create a new problem inthat some customers do not know how to use the virtual order list. Thiscan then be solved with some help functionality on the mobile phone.
Some customers prepared the shopping list from home By having a virtual cata-logue of all the available products on the mobile phone, the customers areable to add products from the catalogue to the virtual shopping list fromanywhere in the world.
Some employees experienced that customers had mistyped product numbers The prob-lem is eliminated since the product number is not manually typed by thecustomer, but instead is read from the NFC tag.
Employees experienced that they occasionally mistyped product numbers into the systemThis problem is also automatically solved as the product number is readfrom the NFC tag. However, if the tag contains the wrong/unreadableproduct number it is possible to manually add the product number.
Employees could not always read the customers handwriting on the shopping list Theproblem is eliminated since the physical shopping list does not exist inEmptio.
42
Customers waited approximately 10 minutes while the orders were being processedThis problem cannot be eliminated however the process can be expeditedby removing the interaction with the store clerks and paying directly fromthe phone. This eliminates the queue to the checkout counter, which inturn can shorten the waiting time. However, this is not a guarantee sincemany factors are in play.
Overall Conclusion
In the System Design chapter we gave a description of how a mobile applicationcan be designed. Even though the Comprehensible quality goal was rated as lessimportant, an effort was made to make a good design. We used object-orienteddesign to structure and describe our system under development, which we choseto call Emptio. As all traditional object-oriented design, we designed objects tomodel concepts from the real world and tried to make these concepts useful andeasy to understand. In real life we have a concept of a shopping list, which wein the design represent as a virtual order list, with its properties and functions.The same applies for other concepts from real life such as order and product.They all have their own object representation in our design.
A good design is also a simple design that is easy to understand and also relies onwell-proven design ideas. This was accomplished by the use of design patternswhich are a general description of a problem and a solution for that problem.Part of the object-oriented design builds on the idea of reuse, and reuse in thiscase are design patterns. We have kept the design simple, as it gives a betteroverview of the system. Furthermore, it does not use over generalization. It isalways possible to make a more general version of a class. For instance in ourcase, the Customer could be derived from a more general class Person, but sinceit does not add any value it was omitted.
We came up with a good design which presents some solutions for the problemsin self-service shopping found in the field study. This was one of the researchquestions raised in the introduction and the design will be implemented andtested next semester.
Limitations
A limitation in the interviews was that we only had seven interviewees. Thisis a very small segment of the customer base in Harald Nyborg and thereforewe might have missed problems that exist in the shopping practice. Also, theseverity of the discovered problems can not be determined.
In the focus group we also had some limitations. The focus group was heldduring a lunch break, which meant that the employees might not have been asfocused as they could have been and the number of participants was also quitehigh and perhaps it should have been lower. Furthermore, the store manageralso participated in the focus group which could cause some of the employeesnot to speak up. The subject could have been boring for some of the partici-pants which also might be the reason for the low engagement. Additionally, weonly had 25 minutes to complete the focus group.
43
CHAPTER 6. CONCLUSION
We had no experience with focus groups and did not conduct a pilot test andthis meant that the course of the focus group was not fluent.
A design can always be improved. What we did not do in this project, wasto take a step back and compare our design to other alternatives and similarsystems. Also, we should have looked more at techniques for idea generationand innovation. Even though we do not have an implementation so that wecan evaluate our design, there are techniques such as paper prototyping that wecould have utilised. Maybe it could have sparked a new idea or made a designflaw obvious.
44
7Bibliography
[1] The International Telecommunication Union ITU. Mobile sub-scriptions. http://www.itu.int/ITU-D/ict/statistics/at_glance/KeyTelecom.html.
[4] Near field communications world. Nfc trials, pilots, tests and live servicesaround the world. http://www.nearfieldcommunicationsworld.com/list-of-nfc-trials-pilots-tests-and-commercial-services-around-the-world/.
[5] Near field communications world. Android nfc: Watch google ceo’s in-terview in full. http://www.nearfieldcommunicationsworld.com/2010/11/16/35050/android-nfc-see-google-ceos-interview-in-full/.
[6] C. Enrique Ortiz (Sun Developer Network). An introduction to near-fieldcommunication and the contactless communication api. http://java.sun.com/developer/technicalArticles/javame/nfc/.
[7] NFC Forum. Nfc forum news conference. http://www.nfc-forum.org/news/june06_architecture_and_specs/Technology_Roadmap_News_Conference_Slides_-_June_2006.pdf.
[8] Enrico Rukzio, Karin Leichtenstern, Vic Callaghan, Paul Holleis, AlbrechtSchmidt, and Jeannette Chin. An experimental comparison of physical mo-bile interaction techniques: Touching, pointing and scanning. Proceedingsof the Eighth International Conference on Ubiquitous Computing, 2006.
[9] SmartTouch project. Smarttouch - interaction as simple as touch. http://ttuki.vtt.fi/smarttouch/www/?info=intro.
[10] Ernst Haselsteiner and Klemens Breitfuß. Security in near field communi-cation (nfc). Workshop on RFID Security 2006, 2006.
[11] ZDNet. Around the world in nfc and con-tactless payments. http://www.zdnet.com.au/around-the-world-in-nfc-and-contactless-payments-339285175.htm.
[14] Giesecke and Devrient. Touch & travel. http://www.gi-de.com/portal/page?_pageid=44,149863&_dad=portal&_schema=PORTAL.
[15] Near field communications world. O2’s homecare solution.http://www.nearfieldcommunicationsworld.com/2009/06/05/31279/nhs-tests-nfc-for-home-care-visits/.
[16] Roy Want, Kenneth P. Fishkin, Anuj Gujar, and Beverly L. Harrison.Bridging physical and virtual worlds with electronic tags. Proceedings ofCHI ’99, 1999.
[17] Dr. Jonathan Lazar, Dr. Jinjuan Heidi Feng, and Dr. Harry Hochheiser.Research Methods in Human-Computer Interaction. Wiley, 2010.
[22] Irene Cappiello, Stefano Puglia, and Andrea Vitaletti. Design and initialevaluation of a ubiquitous touch-based remote grocery shopping process.Proceedings of the 2009 First International Workshop on Near Field Com-munication, 2009.
[23] Panos Kourouthanassis and George Roussos. Developing consumer-friendlypervasive retail systems. IEEE Pervasive Computing archive Volume 2,Issue 2, 2003.
[24] Sangeetha Shekar, Prashant Nair, and Abdelsalam (sumi) Helal. igrocer aubiquitous and pervasive smart grocery shopping system. Proceedings ofthe 2003 ACM symposium on Applied Computing, 2003.
[25] Darren Black, Nils Jakob Clemmensen, and Mikael B. Skov. Supporting thesupermarket shopping experience through a context-aware shopping trolley.OZCHI ’09 Proceedings of the 21st Annual Conference of the AustralianComputer-Human Interaction Special Interest Group, 2009.
The research community has shown an increasing interest in using ubiquitoustechnologies to support our shopping activities. Improving customers experienceby using emerging technologies is the next phase in the mobile revolution. Thefollowing paragraphs shows research in this area.Irene et al. [?] seeks to trigger remote purchasing of groceries via new forms ofHCI in ubiquitous computing. They have conceived a remote grocery shoppingprocess based on touch enabled technologies. A customer can prepare a shop-ping list on an NFC enabled phone by touching on RFID tags attached on items.He opens the refrigerator, touches the RFID tag on the remaining corn flakes,enters wanted quantity and touches the "buy" icon on the fridge and the orderis sent. The results of their experimental research shows that touch-based so-lutions are highly intuitive and users perceived it as user-friendly and satisfying.
In another paper, two researches are primarily focusing on consumer percep-tions of new services and applications. They developed a pervasive retail sys-tem called MyGrocer [?]. The idea is, that a customer picks up a shopping cartwith a tablet PC and an RFID sensor, which detects products placed in thecart. Every time a product is added to the cart, it is automatically removedfrom the shopping list on the screen. The system can show description for aproduct, detailed information, total cost of cart’s contents, provide navigationassistance and other features. Participants in the evaluation agreed that usingthe system reduced their stress levels and they believed it would improve theirshopping experience. Constant awareness of total cost, price comparison, navi-gation system, and smart checkout, where the features that participants foundmost attractive. iGrocer is another attempt to utilise pervasive computing tech-nologies in grocery shopping [?]. This smart grocery shopping assistant makesuse of a mobile phone with a bar code scanner accessory to provide personalisedshopping lists, nutrition profile, "add by recipe", and other features similar toMyGrocer.
Similar to MyGrocer, CAST (Context-Aware Shopping Trolley) seeks to enhancethe shopping experience in supermarkets [?]. It can show product placement inthe store and provide additional information about the individual products. Asthe customer moves around in the store, the system is reordering the shoppinglist based on the location of the customer and the proximity of items. When
the customer is nearby a product, the system sends a audio notification throughthe customer’s headset.
While supermarket shopping is an essential and routine type of activity, de-signing ubiquitous systems that can support the shopping activity and enhancethe experience has received substantial interest. This is evident in the numberof research papers, that concerns the topic. The research conducted confirmsthe promise of user-friendly, intuitive, and robust form of interaction on NFCenabled ubiquitous systems.
49
BCustomer questionnaire
Before the observation begins, ask the customers about following thing:
• Age:
• Name:
• Gender:
• How often do you shop in Harald Nyborg?:
During the observation ask the following questions:
• What do you think of the current shopping process i Harald Nyborg? (Manually entering product ids on a shopping order form )
• Have you experienced any problems with the current shopping process?
• Have you any ideas for improving the current shopping process?
• Do you have a mobile phone? ( If yes ) Is it a smartphone?
• What do you use it for? To call others, texting, MMS, browsing theInternet, taking pictures?
• Can you imagine that a handheld device could be used for shopping? (mobile phone, PDA or similar )
CFocus group agenda
Agenda:
1. Introduction to the focus group (Explain about our education and thatwe are interested in using technology to improve the shopping experience.
2. Questions for the focus group
• Do you experience problems with the current system?
– Both as customer and employee
• Do you experience problems with the customers?
– read the customers handwriting– customers mistype on the shopping list– customers do not know how to use the shopping list
3. Questions for "Guided Fantasy"
• If you could decide freely, how what would should the system be ableto do?
• How would the screen look in certain situations?
– When you place a product in the shopping basket– When you place your order– When you click the shopping basket
• How would you be able to remove products from the shopping basket?
4. Explain our idea for the product
If the focus group stalls give them some topics to talk about:
• Should the customers be able to rate products?
• Should the customers be able to see related products?
• Should the customers be able to see an order history?
• Should the customers be able to see more specific info about products?
• Should the customers be able to see if the products are in stock?
• Should the customers be able to navigate around the shop with the system?
51
APPENDIX C. FOCUS GROUP AGENDA
• Should the customers be able to see a catalog of all the products in thestore?
• Should the customers be able to search for products?
Note: Give the participants a smartphone so they can imagine the system onit
52
DFocus group transcription
[Rasmus] Jeg er Rasmus og det er TorVi er igang med et specialeprojekt hvor vi gerne vil forbedre indkøbsprocessenpåden måde I har indkøbsprocessen her hvor I har folk, der selv går rundt ogskriver hvad de skal have af varer.Selvfølgelig tager de ogsåselv, men vi vil forbedre den indkøbsproces med teknologiog vi har tænkt lidt over sådan noget mobilteknologiOg sådet vi egentlig gerne vil have ud af jer er input til idéer til hvordan detskal foregåog det gør vi med noget der hedder Guided Fantasy. I får egentlighelt selv lov til at komme med idéer til hvordan jeres system skal forbedres.[HN medarbejder] Hvordan fungerer det?[Rasmus] Jamen, det fungerer egentlig.. alt efter hvordan.. de idéer vi får udfrajer.. Jeg kommer med noget input og såvil jeg egentlig gerne høre I kunneforestille jeg det skulle fungereNu kender I jeres eget system rigtig rigtig godt, går jeg vel udfra, nu når Iarbejder her [latter] Og såderfor er vi.. det vi gerne vil gøre er at, hvis detskulle komme op at køre det her, såskulle det selvfølgelig være noget som derikke ville være nogen påvirkning for jer såI ville køre det samme, påden sammemåde sådet egentlig kun ville være kunderne der ville tage deres telefon og brugeden istedet for for at handle ind med.Sådet skulle egentlig fungere.. et eksempel kunne være at man havde sin telefonsom såkunne scanne de der numre istedet for. Sådu fik bare en liste påtelefonenover de varer du havde og såkunne man trykke (køb) og såville I fådet indpåskærmen istedet for de kom med listen, som et eksempel.Men, det vi egentlig gerne ville vide det var om I har nogen problemer... Ioplever nogen problemer som det er lige nu, med det nuværende system.. altså..ikke at systemet laver fejl, men at når kunden kommer og afleverer sedlen at Iikke kan læse hvad der står.[HN medarbejder] Det er tit at de komme med forkert bestillingsnummer[Rasmus] Ja[HN medarbejder] At vi ikke kan læse hvad de skriver[HN medarbejder] Især det.. har du ret i[HN medarbejder] Kæmpe fordel, hvis man kan undgåsådan noget[Rasmus] Ja, sådan at der ikke er den fejl med at det er kundens håndskrift..eller at de kan skrive forkert[HN medarbejder] Ja, lige præcis[Rasmus] Og så, hvad mere er der.. jo eh, I oplever heller ikke noget internt, nårI sender noget udefra disken og såud pålageret.. der opstår heller ikke noget fejl
53
APPENDIX D. FOCUS GROUP TRANSCRIPTION
i kommunikationen eller noget derimellem fordi det hele kører over systemet,eller noget[HN medarbejder] Nej[HN medarbejder] Nej.. ikke andet end noget paperjam [latter][Rasmus] Hvad med .. er der nogen kunder der kommer op og ikke kan finde udaf at bruge den her seddel, eller?[HN medarbejder] Ja.. nye kunder, ja[Rasmus] Altsåde komme op og skal sådan en introduktion til hvordan man gårrundt og bruger den[HN medarbejder] Ja, det kommer.. når konen skal med at handle.. [latter][Rasmus] Det var det..Og såer der sådan at vi godt kunne tænke os at I egentlig bare har frie tøjler tilat.. hvis man kunne forestille sig at man havde sådan en telefon.. behøver ikkeat være en smartphone [lægger telefonen påbordet] det kunne ogsågodt barevære en almindelig telefon. Når man sågik hen og såscannede sine varer indeller om man bruger camera til at læse nummeret der står eller der er sådan enlille sender I .. enten med bluetooth eller noget andet såman kunne bare gåhenog sige "bip" når man holdt den op imod varen og fik den ind påskærmen.. hvadman såville fåaf information påskærmen.. Du måogsågodt komme med inputhvis du har noget..[til Niels Christian].. hvad man kunne fåaf informationer nårman fik sådan en vare ind.. skal der være billede med.. skal der være ekstratekst.. ligesom I har i jeres katalog.. der står lidt ekstra til nogen af varerneend hvad der står ude ved selve varen.. sådan nogen ting.Hvad kunne I forestille jer der var ved det her billede.. det her skærmbillede nårman nu scannede sine varer ind?[HN medarbejder] Tror ikke det skulle være for meget til de kunder vi har fordidet ville forvirre dem alt for meget hvis der komme for meget frem[Rasmus] Ja[HN medarbejder] Vi har jo mange ældre[Rasmus] Ja.. det er jo lige det.. vi ved jo ikke hvordan kunderne er herudeudover det vi har fået fortalt sådet er det vi ville høre fra jer hvordan I opleverdet skal være for kunderne.[HN medarbejder] Altså.. jeg tror ikke det skal være alt for teknisk for såvil detforvirre dem og såer de væk fordi dét der det kan jeg ikke finde ud af og så..[Rasmus] Såe.. ligesom det er herude at bare navnet påvaren, og et billede afvaren, for eksempel.[HN medarbejder] Ja[Rasmus] Og såen pris.. såholde det såsimpelt som muligt[HN medarbejder] ja.. det tror jeg[Niels Christian] Kunne det ikke være smart nok hvis de fik samme informationsom der står i vores hovedkatalog? De behøves jo ikke at fåden vi har i voresonline [...][HN medarbejder] Det tror jeg da ville besvare spørgsmål, både ude pågulvetog oppe i ordre-[...][HN medarbejder] Sådan et sted som [...] tror jeg ville være fint nok [....][HN medarbejder] Men det er altid en balancegang..[HN medarbejder] Det har du fuldstændig ret i.. Det er ogsånoget der bliverarbejdet meget med varemæssigt fra indkøberens side af at vi forsøger påatfåvores kunder yngre og yngre. ogsåselvom vi sælger rollatorer.. [latter].. vi skaljo ramme alle, ikke også?.. men der er hvertfald meget fokus indkøbsmæssigt
54
[Rasmus] Man kan såogsåsige at vores produkt det skal heller ikke blive noget derkommer ud i morgen.. det er teknologien heller ikke moden til.. sådet er sådannoget der skal lige gåen årrække af 2-5 år eller sådan noget før det begynder atvære modent nok og folk de måske ogsåbegynder at turde at bruge denne herteknologi. Men hvad .. sånu vender vi lige tilbage til det der med skærmbillede..hvad nu hvis man forestiller sig at man har den der liste af ting man har købt... hvordan kunne man forestille sig at man skulle bruge telefonen til at købemed.. om man skulle gåop til kassen og såvise den eller skulle man bare haveen knap hvor man trykker "send" og såkøbte man og skulle man såbetale oversit abbonement eller har I nogen idéer..[HN medarbejder] Der kunne være sådan et nummer påselve telefonen hvor visåslår det nummer ind påskærmen.. det kunne man..[Ramus] JaI har også.. hvordan er det systemet herude når kunderne kommer.. får desådan et nummer som I kalder op.. eller har I et skilt derude?[HN medarbejder] Der kommer såsådan et nummer når de afleverer sedlen[Rasmus] Ja.. okFordi såkunne man såogsåforestille sig at det nummer det fik man bare påskær-men, og så..Hvad såmed.. de her ting.. når man.. altså, når man tager noget inde påskær-men såskal man selvfølgelig ikke købe det med det samme altså.. scanner.. detkan jo godt være at man bare ville have informationen påproduktet.. såom derskal være.. hvordan man såskal lægge det over i sin indkøbskurv .. hvordan denskal se ud.. hvis I kunne forestille jer det. Jeg ved godt det er meget abstraktat sidde og forestille sig det ligenu, men hvis I kunne forestille jer at man fikinformation og der var et lille billede og tekst påhvad det nu var man scannedeind.. opvaskemaskine.. og såen pris.. og såen lille knap hvor man kunne sige.."Køb" eller et eller andet.. Når man såtrykker påden såfik man en indkøbskurvfrem med alle de ting man havde.. skulle det bare være en lang liste eller kunneman forestille sig det skulle se ud påen anden måde såman havde bedre overblik[HN medarbejder] Nej.. det skulle bare være en liste[Rasmus] Jeg ved godt det er meget abstract [latter]Men det er nu sådan noget som det vi havde før.. det er sådan noget vi gernevil vide.. I har noget intern information og det er det vi gerne vil have ud ogdet er selvfølgelig svært for jer at vide hvad for noget vi gerne vil vide og det erogsåsvært for os at fåstillet de rigtige spørgsmålHvad med sådan noget at lade kunderne komme med kommentarer til produk-terne.. det ved jeg ikke om I har inde påhjemmesiden[HN medarbejder] [latter] der er nogle enkelte kunder der kommer med kom-mentarer [....][Rasmus] Men.. sådan at når kunderne.. hvis de har købt et produkt påettidspunkt og de kunne tage programmet med hjem påtelefonen og såkunne dese hvad de havde købt og sågive .. evt. give kommentarer til et produkt ellerkomme med stjerner alt efter hvor godt det nu virker eller noget.. om det kunnevære en ide[HN medarbejder] [mange samtidigt] Nej.. det tror jeg ikke[Niels Christian] Altså.. langt de fleste af dekunder der handler herude hos os.. såringer de ikke for at sige at den skruetrækker den var fandme go’.[Rasmus] Det er fordi at vi oplevede meget da vi stod derude og snakkede meddem at der var rigtig mange der kommer.. det er ikke førstegangs købere.. det
55
APPENDIX D. FOCUS GROUP TRANSCRIPTION
er folk der kommer her rimelig ofte.. sådan en .. en to gange om måneden,hvertfald. Det måvære folk der er glade for at komme tilbage her at købe[HN medarbejder] Jaja.. selvom de er negative såkommer de alligevel[Rasmus] Det er jo ogsåaltid nemmere at være negativ end det er at kommemed det gode feedbackDet er mere det at det kunne være en mulighed.. men det lyder ikke som omdet er der skal være med eller..[HN medarbejder] [kan ikke tyde hvad personen siger][Ramus] Det er såogsånoget man kunne forestille sig at kunne hjælpe med ven-tetiden da folk ikke skal op i køen for at aflevere de der ting.. men at de barekunne[HN medarbejder] ja.. at de køber ved at bare indtaste de her fire cifre påtele-fonen[Rasmus] ja.. eller ogsåat de eventuelt kunne købe tingene undervejs som definder dem.. at man kunne sige at de kunne lave en .. de finder den skruetrækkerde skal have men de ved ogsåat de skal ned og en opvaskemaskine.. sånår de harfundet skruetrækkeren og scanner den ind sålægger de bestillingen allerede derpåskuretrækkeren som egentlig er klar såskal de såud at finde vaskemaskinennår det er den der skal bestilles.. sådan at varen er fundet undervejs.. altsådetkunne selvfølgelig godt give..[Niels Christian] Nej.. jeg tror ikke.. ordren skal findes påén gang for at viskal kunne holde styr pådet der.. vi prøvede noget lignende pået tidspunkt iOdense hvor kunderne kunne bestille varerne pånettet til afhentning i Odense..det skulle vi såfinde ud af om det var noget var aktuelt for os [....].. det gikrigtig godt.. ca hver femte ordre blev hentet.. resten kunne de lægge tilbagepåplads[Rasmus] Ok, såkan jeg godt se det..[Niels Christian] Der var hvertfald rigtig meget arbejde i det... det kørte alligeveli et par måneder[Rasmus] Og sålige for at runde lidt af.. såvores ide. altsåsådan den færdige idetil produktet.. det er at bruge en teknologi der hedder NFC.. som er en trådløsstandard der kan læse småchips.. jeg ved ikke om I har hørt om RFID chips?Der er egentlig bare småchips der koster 20 øre stykket eller sådan noget somman kan sætte i alt og såkan det indeholde en masse information.. og dem kanman såaflæse med forskellige apparater.. og det er såegentlig det der er vores idé.. at kunderne kunne komme med deres mobiltelefoner da fordi fra 2011 skulleder begynde at komme NFC i alle mobiltelefoner og såkan de komme og aflæseet produkt med telefonen og fådenne her information og lave deres indkøbslisteog såbestille.. hvis det såkommer til at virke ordentligt .. såkunne man forestillesig engang at det kunne integreres i sådan et system som I har her.. fordi det eren rigtig god case.. fordi I kører det påden måde.. og sågøre indkøbsprocessennemmere for kunderne og såsamtidig ogsågøre bestillingsprocessen nemmere forjer at der ikke er ligesåmange fejl sådet måske kan gøres hurtigere[Niels Christian] Ja vi laver jo ogsåtastefejl når vi står oppe ved ordre.. når vitaster.. det kan jo ikke undgås[Rasmus] Man kunne ogsåforestille sig at kunderne kunne se lagerstatus påvar-erne de vil købe og fåat vide om den er her eller ej..[Niels Christian] Ja.. det har vi sålidt politik i at det fortæller vi ikke.. alt-såkunderne.. der er mange der ringer herud for at spørge.. har I den eller den
56
pålager.. hvor mange har du.. såspørger man hellere om hvor mange de vilhave.. der får man ikke at vide om man har 7 eller 77[Tor] Her tænker vi bare påom I har varen pålager eller ej og ikke om I har 17tilbage[Rasmus] AltsåI har.. I kan godt se hvor mange der burde være herude[Niels Christian] Ja[Rasmus] Det er bare en lille grøn pære om den er der eller eller om den ikke erderBare sådan at man ikke går rundt i butikken og finder 3 ting der ikke er der ogsåstår i køi lang tid for at finde ud af at det kan man ikke fåi dag så[HN medarbejder] Det er jo risikoen[Rasmus] Nu ved jeg ikke om I har styr på.. altsåI har de her medlemskort.. erder såmulighed for at se..[Niels Christian] Hvad er det for nogen kort?[Rasmus] Det ved jeg ikke.. Det var noget.. en brochure jeg såoppe ved kassen[Niels Christian] Ahh.. det er kunde[...].. det bliver ikke brugt særlig meget[Rasmus] Ok[Niels Christian] Det er med en rente på28-29 %.. såer nordjyder sgu ikke såvildemed det[Rasmus] Ok.. ej ok.. det kan jeg godt se.. [latter]Jeg såden bare lige og såtroede jeg det var sådan et medlemskort[Niels Christian] Det er egentlig vores .. ej det har været vores eget kort.. menvi valgte at udlicitere dem da der var for store tab pådem..Såumiddelbart er ikke den kundegruppe vi er mest interesseret i.. jo vi vil gernehave dem ud.. nu har vi ikke selv tab pådem.. det er meget meget lille[Rasmus] Det var egentlig bare det vi gerne ville vide.. jeg siger tak for hjælpenog vil lige høre om I lige gider tage sådan en her og lige skrive navn og hvad dernu ellers er på
57
EShopping lists from Harald Nyborg and
IKEA
Figure E.1: Self-service shopping list from IKEA.
Figure E.2: Self-service shopping list from Harald Nyborg.
